How to back test?
Before you put your real money in a new trading strategy you wanna see how it works, don't you. That was great if we could see the exact results - in advance. That could spare us lots of losing trade, save time and money. I believe that no serious trader invest their money without testing the investment (trading) strategy because they want to know their realistic expectation.
The most obvious solution could be the backtest. When you have a trading idea, a strategy you look at back on the chart for example a year and see how it worked in the past. This way you can have a good conception about the numbers of signals within a period of a year and its reliability.
If you are creative enough and put more effort in it you can write a script in a charting software and run that automatically - that's a little bit more advanced and effective way: the script won't miss any of the signal by mistake and give you detailed statistics about winning and losing trades, their proportions, etc. Doing this your script will check the chart for that particular signal from the past to today and pretend to trade all of them. This opportunity presents itself to see the trades and its results as you traded them. If you are still not happy with the outcome you can modify or fine tune your strategy and run the script again to see the differences. It's an exciting and interesting 'game' to see how a small modification can cause a huge effect for the final outcome. Once you finished optimalising your strategy you are ready to put in practice. Are you really ready for that?
Past performance is no guarantee of future returns!
During your back test procedure you probably got an average of 35% return on a monthly basis. Unfortunately and surprisingly when you start trading with that you experience much less or even losing series. You give it a couple of more chances and after a number of losses you finally give that up and start backtesting a new strategy. Familiar?
Back test and its deficiencies
Back testing is a two edged tools. It can spare you time and hard work but gives you false results - there is no exception it is always cheating you.
Market is always changing - we all know that market conditions like liquidity and volatility are always and continouosly changing. Because of this reason the past performance is no garantee of future returns.
Back test never sleeps - back test script trade all of the signal that match your criteria. It trades when you sleep, when you are with your family; all day and every day. Obviously you are not able to do that so you are not able to trade that number of trades. Also, (on the Forex Market) it trades all of the markets: London, New York, Singapore, Tokyo, Sydney, Frankfurt. I can tell you that no strategy can work in all markets as they have very different conditions providing very different trading opportunities. If you are unlucky and trade in that session which doesn't provide you proper market conditions - you will lose. (eg. Sydney session is obviouosly NOT the best period for trading EURUSD because of the lack of its liquidity - which comes in London session.)
Back test doesn't consider the news - you will find many trading possibilities and signals that you can't trade because of the imminent economical news. News usually provides extra volatility on the chart and seeing them afterwards indicates clear signals. Unfortunately trading them in time would not more than gambling - because we can't determine the results of the news and its effect for the market in advance.
Back test doesn't suffer from psychology - when you've got 3 consecutive losses (it can happen) your hands most likely will be shaking opening a new position when a new signal comes. Back test doesn't care the losses or winnings; it just trades.
Back test can survive drawdown - although your strategy performs very well in a certain period, it might has a very bad drawdown which causes 50% loss or even more. Because it is just a script and can indicate negative balance as well, it will continue working and building up the account again. That is what you CAN NOT afford and will stop trading around 20-30% losses; that's for sure. No one can trust any strategy which causes 30% loss in a given time period and says: ' That's ok, no problem let's do it again.' Moreover you are financially not able to continue trading as your real broker won't let you trading in negative balance, will they.
Which way can I test properly?
Because of the reasons above I developed my Real Time Testing System to eliminate those problems. At the moment I am writing these lines we work more than 7000 logged trades and have a massive statistics. Before I open a certain position I exactly can see its reliability under the specific circumstances ( currency pair, time window, time frame, market open, etc.) so I am very confident to trade that or not. Obviously I can't be deadly sure but I can reach as high as 80-85% of reliability with this way - and it's more than enough for confident and lucrative trading.


